Texture enhanced underwater image restoration via Laplacian regularization

Hao, Yali; Hou, Guiting; Tan, Lifeng; Wang, Yongfang; Zhu, Haotian; Pan, Zhenkuan

doi:10.1016/j.apm.2023.02.004

Cited by 7 publications

(1 citation statement)

References 48 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…To estimate model parameters, some scholars have used deep learning networks to estimate background light and transmission map or scene depth, which relies on the network structure design and training data. In addition, Hao et al [ 10 ] developed an underwater laplace variational model and used luminance mixing and quadratic tree subdivision algorithms to estimate the transmission map and background light. Xie et al [ 11 ] proposed a red channel prior guidance variational framework, which successfully combined the normalized total variational term and the sparse prior knowledge of the fuzzy kernel to achieve better underwater image enhancement results.…”

Section: Section 2: Related Workmentioning

confidence: 99%

Underwater image enhancement using Divide-and-Conquer network

Zheng,

Wang,

Chen

et al. 2024

PLoS ONE

View full text Add to dashboard Cite

Underwater image enhancement has become the requirement for more people to have a better visual experience or to extract information. However, underwater images often suffer from the mixture of color distortion and blurred quality degradation due to the external environment (light attenuation, background noise and the type of water). To solve the above problem, we design a Divide-and-Conquer network (DC-net) for enhancing underwater image, which mainly consists of a texture network, a color network and a refinement network. Specifically, the multi-axis attention block is presented in the texture network, which combine different region/channel features into a single stream structure. And the color network employs an adaptive 3D look-up table method to obtain the color enhanced results. Meanwhile, the refinement network is presented to focus on image features of ground truth. Compared to state-of-the-art (SOTA) underwater image enhance methods, our proposed method can obtain the better visual quality of underwater images and better qualitative and quantitative performance. The code is publicly available at https://github.com/zhengshijian1993/DC-Net.

show abstract

Section: Section 2: Related Workmentioning

confidence: 99%